physical aspects of polymer-carbon nanotube interactions ... · prof. melissa a. pasquinelli...
TRANSCRIPT
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
Physical Aspects of Polymer-Carbon Nanotube Interactions with
MD Simulations and Visual-Analytic Tools
Prof. Melissa A. PasquinelliAssistant Professor
Fiber and Polymer Science Program
Department of Textile Engineering,
Chemistry, and Science (TECS)
North Carolina State University
Instant Messaging: DrPasquinelli(Yahoo, AIM, Google, MSN)
Phone:(919) 515.9426
http://www.te.ncsu.edu/mpasquinelli
Graduate Student:Syamal Tallury
(FPS/MSE)
Funding:Free software (VMD, LAMMPS, PackMol)
Collaborator:Sid Thakur,
RENCI@NC State
www.renci.org
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
Non-Covalent Interactions with Polymers and CNTs
Interaction depends on:
Structure and functionalization of CNTs
Structure of polymeric surroundings
Interfacial interactions between the polymer
chains and CNTs
Polymer “wrapping”
can be utilized to:
•Solubilize CNTs
•Effectively disperse the
nanotube reinforcement
in the matrix of a
nanocomposite
•Alter the properties of
the material
Presence of CNTs can
impact properties of
the material:
•Toughness
•Mechanical strength
•Crystalline morphology
•Conductivity
•Biocompatibility
M Naito, K Nobusawa, H Onouchi, M Nakamura, K-I Yasui, A Ikeda and M Fujiki; JACS 2008, 130 (49), 16697
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
RESULTS: Polymer-CNTs in Vacuum
Syamal Tallury
PP
PEO
N6PCLPLA
PAN
PS
PMMA
SS Tallury and MA Pasquinelli, JPCB 114(12): 4122–4129 (2010); JPCB 114(29): 9349–9355 (2010).
PA
PET
PPEPPYPPV
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
PA-5 multichains PA-10 multichains
• Observe structure-property relationships • Among polymer molecules of same type and different types
• Identify patterns in spatial and temporal trajectories• Could provide insight into properties of interest, for instance the induction of polymer crystallization within an amorphous media
• Explore relationships among system parameters• Understand what dictates the dynamics of the interfacial interactions of polymer-nanoparticle
• Provide a tool for making rational choices about system parameters• To improve the microstructure and the macroscopic properties of a material by tuning quantities such as the chemical composition or processing conditions
Visual-Analytic Tools for Particle Simulation Data
SA Thakur and MA Pasquinelli, Macromolecular Theory and Simulations (2011) 20(4): 286-298. SA Thakur, SA Thakur, SS Tallury, MA Pasquinelli, and TM Rhyne, in Advances in Visual Computing (2009) 129-139.
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
Comparing polymer conformations at distinct time steps
‘
C. Best and H.-C. Hege, “Visualizing and identifying conformationalensembles in molecular dynamics trajectories,” Computing in Scienceand Engg., vol. 4, no. 3, pp. 68–75, 2002.
Conformation refers to the 3D structure of a polymer molecule in a single time step
Basic Approach
SA Thakur and MA Pasquinelli, Macromolecular Theory and Simulations (2011) 20(4): 286-298. SA Thakur, SA Thakur, SS Tallury, MA Pasquinelli, and TM Rhyne, in Advances in Visual Computing (2009) 129-139.
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
di = | p(i+1) – p(i) |
Example of a numeric measure: Atomic distances
t
t = time step
[Best and Hege, 2002]
p(i-1)
p(i)
p(i+1)
di
polymer backbone bonds and atoms at time step `t’
A feature vector encapsulates some property of a polymer conformation using a numeric measure.
Feature Vector
SA Thakur and MA Pasquinelli, Macromolecular Theory and Simulations (2011) 20(4): 286-298. SA Thakur, SA Thakur, SS Tallury, MA Pasquinelli, and TM Rhyne, in Advances in Visual Computing (2009) 129-139.
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
• Compute correlations of polymer conformations
1
2
Methodology
SA Thakur and MA Pasquinelli, Macromolecular Theory and Simulations (2011) 20(4): 286-298. SA Thakur, SA Thakur, SS Tallury, MA Pasquinelli, and TM Rhyne, in Advances in Visual Computing (2009) 129-139.
Data is normalized relative
to the largest RMSE
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
APPLICATION: Polymer-CNT Interactions
SA Thakur and MA Pasquinelli, Macromolecular Theory and Simulations (2011) 20(4): 286-298. SA Thakur, SA Thakur, SS Tallury, MA Pasquinelli, and TM Rhyne, in Advances in Visual Computing (2009) 129-139.
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
Examples of Numeric Metrics
SA Thakur and MA Pasquinelli, Macromolecular Theory and Simulations (2011) 20(4): 286-298. SA Thakur, SA Thakur, SS Tallury, MA Pasquinelli, and TM Rhyne, in Advances in Visual Computing (2009) 129-139.
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
Similarity Matrices with Different Numeric Metrics
SA Thakur and MA Pasquinelli, Macromolecular Theory and Simulations (2011) 20(4): 286-298. SA Thakur, SA Thakur, SS Tallury, MA Pasquinelli, and TM Rhyne, in Advances in Visual Computing (2009) 129-139.
Data is
normalized
relative to the
largest RMSE
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
Indeces for Similarity Matrices
PP PA
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
Matrices using a segmented color to display similarity relationships between conformations of Nylon-5 during an MD simulation in two distinct cases, namely (a) with a CNT, and (b) without a CNT.
APPLICATION: Polymer-CNT Interactions
SA Thakur and MA Pasquinelli, Macromolecular Theory and Simulations (2011) 20(4): 286-298. SA Thakur, SA Thakur, SS Tallury, MA Pasquinelli, and TM Rhyne, in Advances in Visual Computing (2009) 129-139.
Data is
normalized
relative to the
largest RMSE
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
Connect to Properties: Atom-Time-Value (ATV) Plots
SA Thakur and MA Pasquinelli, Macromolecular Theory and Simulations (2011) 20(4): 286-298. SA Thakur, SA Thakur, SS Tallury, MA Pasquinelli, and TM Rhyne, in Advances in Visual Computing (2009) 129-139.
Examples of “values” or measures:
• Distance of backbone atoms from some origin (such as center of mass)
• Instantaneous velocities of backbone atoms
• Relative displacements of backbone atoms with respect to an initial set of
locations (pseudo persistence length)
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
APPLICATION: ATV Plots for Polymer-CNT Interactions
Without Carbon Nanotube:
With Carbon Nanotube:
SA Thakur and MA Pasquinelli, Macromolecular Theory and Simulations (2011) 20(4): 286-298. SS Tallury and MA Pasquinelli,
JPCB 114(12): 4122–4129 (2010); JPCB 114(29): 9349–9355 (2010).polyacetylene
z
y
x
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
Other Visualization Aspects: Linking and Brushing
SA Thakur and MA Pasquinelli, Macromolecular Theory and Simulations (2011) 20(4): 286-298. SA Thakur, SA Thakur, SS Tallury, MA Pasquinelli, and TM Rhyne, in Advances in Visual Computing (2009) 129-139.
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
Other Visualization Aspects: Projections
SA Thakur and MA Pasquinelli, Macromolecular Theory and Simulations (2011) 20(4): 286-298. SA Thakur, SA Thakur, SS Tallury, MA Pasquinelli, and TM Rhyne, in Advances in Visual Computing (2009) 129-139.
polyacetylene
z
y
x
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
Pattern Matching in Spatial and Temporal Dimensions
SA Thakur and MA Pasquinelli, Macromolecular Theory and Simulations (2011) 20(4): 286-298. SA Thakur, SA Thakur, SS Tallury, MA Pasquinelli, and TM Rhyne, in Advances in Visual Computing (2009) 129-139.
Need to simplify
the molecular
(particle)
structure in
spatial
dimensions
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
Substructure Matching
• Algorithm:
– Divide arc length curve into
multiple segments
• Segment size set by user
– Compared with RMS Error
between feature vectors
• Tolerance controlled by
user
– Performed for all possible
unique pairs of curve
segments
• Output:
– ID most commonly occurring
features within polymer
conformations among set of
related polymers
• Series of nylons
– Explore spatial and temporal
distributions of specific features
• Start of crystal formation
– Compare global structures of
related polymers or under
different scenarios
• With/without a nanoparticle
present
SA Thakur and MA Pasquinelli, Macromolecular Theory and Simulations (2011) 20(4): 286-298. SA Thakur, SA Thakur, SS Tallury, MA Pasquinelli, and TM Rhyne, in Advances in Visual Computing (2009) 129-139.
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
Other Visualization Aspects: Clustering with MDS
SA Thakur and MA Pasquinelli, Macromolecular Theory and Simulations (2011) 20(4): 286-298. SA Thakur, SA Thakur, SS Tallury, MA Pasquinelli, and TM Rhyne, in Advances in Visual Computing (2009) 129-139.
Multidimensional Scaling:
High dimension (N) to
low dimension (d, d << N)
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
Other Details
• Stand-alone program
• Works with LAMMPS MD data sets
– Will be tested on DPD soon
• Currently being extended to multi-molecule systems
(polymer melts, solvent systems, etc.)
• Will be publicly available once multi-molecules are
working (hopefully by end of the year)
– On my website: www.te.ncsu.edu/mpasquinelli
– On RENCI’s website: www.renci.org
– On LAMMPS website
http://www.te.ncsu.edu/mpasquinelliTHE LABORATORY OF MULTI-SCALE MODELING FROM THE NANOSCALE
Contact Information
Instant Messaging: DrPasquinelli(Yahoo, AIM, Google, MSN)
Phone:(919) 515.9426
http://www.te.ncsu.edu/mpasquinelli
Funding:
Free software (VMD, LAMMPS, PackMol)
NC-ACS
Project
SEED